In this paper, feature extraction methods are developed based on the non-negative matrix factorization (NMF) algorithm to be applied in weakly supervised sound event detection. Recently, the development of various features and systems have been attempted to tackle the problems of acoustic scene classification and sound event detection. However, most of these systems use data-independent spectral features, e.g., Mel-spectrogram, log-Mel-spectrum, and gammatone filterbank. Some data-dependent feature extraction methods, including the NMF-based methods, recently demonstrated the potential to tackle the problems mentioned above for long-term acoustic signals. In this paper, we further develop the recently proposed NMF-based feature extraction method to enable its application in weakly supervised sound event detection. To achieve this goal, we develop a strategy for training the frequency basis matrix using a heterogeneous database consisting of strongly- and weakly-labeled data. Moreover, we develop a non-iterative version of the NMF-based feature extraction method so that the proposed feature extraction method can be applied as a part of the model structure similar to the modern “on-the-fly” transform method for the Mel-spectrogram. To detect the sound events, the temporal basis is calculated using the NMF method and then used as a feature for the mean-teacher-model-based classifier. The results are improved for the event-wise post-processing method. To evaluate the proposed system, simulations of the weakly supervised sound event detection were conducted using the Detection and Classification of Acoustic Scenes and Events 2020 Task 4 database. The results reveal that the proposed system has F1-score performance comparable with the Mel-spectrogram and gammatonegram and exhibits 3–5% better performance than the log-Mel-spectrum and constant-Q transform.

In recent acoustic scene classification (ASC) models, various auxiliary methods to enhance performance have been applied, e.g., subsystem ensembles and data augmentations. Particularly, the ensembles of several submodels may be effective in the ASC models, but there is a problem with increasing the size of the model because it contains several submodels. Therefore, it is hard to be used in model-complexity-limited ASC tasks. In this paper, we would like to find the performance enhancement method while taking advantage of the model ensemble technique without increasing the model size. Our method is proposed based on a mean-teacher model, which is developed for consistency learning in semi-supervised learning. Because our problem is supervised learning, which is different from the purpose of the conventional mean-teacher model, we modify detailed strategies to maximize the consistency learning performance. To evaluate the effectiveness of our method, experiments were performed with an ASC database from the Detection and Classification of Acoustic Scenes and Events 2021 Task 1A. The small-sized ASC model with our proposed method improved the log loss performance up to 1.009 and the F${}_1$-score performance by 67.12%, whereas the vanilla ASC model showed a log loss of 1.052 and an F${}_1$-score of 65.79%.

Monitoring domestic activities helps us to understand user behaviors in indoor environments, which has garnered interest as it aids in understanding human activities in context-aware computing. In the field of acoustics, this goal has been achieved through studies employing machine learning techniques, which are widely used for classification tasks involving sound recognition and other objectives. Machine learning typically achieves better performance with large amounts of high-quality training data. Given the high cost of data collection, development datasets often suffer from imbalanced data or lack high-quality samples, leading to performance degradations in machine learning models. The present study aims to address this data issue through data augmentation techniques. Specifically, since the proposed method targets indoor activities in domestic activity detection, room transfer functions were used for data augmentation. The results show that the proposed method achieves a 0.59% improvement in the F1-Score (micro) from that of the baseline system for the development dataset. Additionally, test data including microphones that were not used during training achieved an F1-Score improvement of 0.78% over that of the baseline system. This demonstrates the enhanced model generalization performance of the proposed method on samples having different room transfer functions to those of the trained dataset.

Acoustic traffic monitoring involves a complex combination of subtasks, including vehicle type classification, vehicle counting, and direction-of-arrival estimation. Effectively training deep learning models for these tasks requires a deep understanding of traffic-related acoustic patterns as well as extensive, accurately annotated datasets, a resource that is limited in public availability. Many existing datasets provide labels for only a subset of these tasks, which complicates the development and evaluation of unified models. While the recent release of the large-scale "Sound of Traffic" dataset by DCASE 2024 Task 10 provided a significant advancement, its size remains insufficient for training complex neural networks from scratch. Consequently, many participants employed a sim-to-real training strategy—pre-training on synthetic simulation data and fine-tuning on real-world recordings. This study proposes an alternative method that leverages discrete audio tokens extracted from a pre-trained DAC to replace conventional mel-spectrogram features in the vehicle type classification pipeline. Our approach achieves competitive performance without relying on simulation-based pre-training, indicating that reliable results can be obtained with reduced development time and computational cost. Notably, experimental results confirm that our system achieves the most stable performance among all evaluated models.